Colorectal cancer screening (CRCS) is usually discussed between the patient and the physician in the context of a clinic visit. However, physicians face a challenge in promoting CRCS in the face of multiple competing demands. A decision aid (DA) that provides decision support and incorporates patient preferences in CRCS test options can aid shared decision making (SDM) and be effective at increasing CRCS rates. However, exactly how such an intervention improves SDM is not clear. Colorectal Web (CW) is a DA that incorporates interactive patient preference clarification and risk assessment that help patients determine the CRCS test option that best matches their preferences. The goal of our study is to provide detailed understanding of how interactive DAs such as CW impact patient-physician communication and SDM, and ultimately CRCS adherence. The central hypothesis is that CW will improve CRCS adherence through improvement in patient behavioral factors, SDM between the patient and the physician, and concordance between the patient's preferred CRCS test and the physician's recommended CRCS test. A 2-armed randomized controlled trial (300 patients per arm) will compare the Intervention Arm using CW to the Control Arm using a non-interactive control website, in 10 practices in Metro Detroit. Patients will be men and women aged 50 years and over, not current on CRCS, and scheduled for check-up or chronic care visit with their physician. In the clinic just before the patient-physician encounter, participating patients will complete a Patient Baseline Survey. They will be randomized to the Intervention Arm or the Control Arm. Data will be collected after the patient review of the respective websites (Post-Web Survey), during the patient-physician encounter (digital audio recording), and after the patient-physician encounter (Patient Post-Encounter Survey). Chart audit will be performed 6 months after the clinic visit to determine whether the patient underwent CRCS (Endpoint Chart Audit). The Specific Aims are: Primary Aim 1: To measure the impact of CW on patient uptake of CRCS. Primary Aim 2: To evaluate the impact of CW on patient determinants, patient preference, and patient intention before the patient- physician encounter. Primary Aim 3: To evaluate the impact of CW on SDM, concordance, and patient intention during and after the patient-physician encounter. Secondary analysis will employ a Structural Equation Modeling approach to understand the mechanism of the causal pathway and test the validity of our proposed conceptual model. The results of our proposed study will be among the first to examine the effect of a real-time preference assessment exercise on CRCS and mediators, and, in doing so, will shed light on the patient-physician communication and SDM black box that currently exists between the delivery of DAs to patients and the subsequent patient behavior. The results will not only have important implications for improving SDM for CRCS but can also be applied to other preference-sensitive care situations where underlying risk factors contribute to screening and/or treatment decisions.